Recently, plagiarism detection company Copyleaks released a report stating that up to 60% of the content generated by OpenAI’s GPT-3.5 model contains different forms of plagiarism. The report raised widespread concerns about copyright and originality issues in AI models. Copyleaks uses a self-developed scoring method that comprehensively considers factors such as text similarity, minor changes, and interpretations. The report shows that computer science, physics, psychology and other subjects have higher similarities, while drama, humanities, English and other subjects have lower similarities. OpenAI responded that it has taken measures to limit unintentional memory of models and explicitly prohibits users from deliberately re-editing content.
The plagiarism detection company Copyleaks recently released a report stating that up to 60% of the content generated by the GPT-3.5 language model launched by OpenAI contains different forms of plagiarism. Copyleaks uses a self-developed scoring method that considers multiple factors such as text similarity, minor changes, and paraphrases to determine similar content as "plagiarism." Computer science, physics, psychology and other subjects have the highest similarity, while drama, humanities, and English have lower similarities. OpenAI stated that it has taken various measures to limit unintentional memory, and the terms of use also clearly prohibit users from deliberately allowing the model to reprogram content. In addition, in addition to the well-known lawsuit filed by the New York Times, some content creators are also trying to claim that the AI model was trained on their works, so the content that produces approximate copies should also be given a share.The report not only highlights the challenges of AI models in content generation, but also triggers discussions on a series of legal and ethical issues such as AI model training data copyright, content originality, and intellectual property ownership. The industry needs to jointly explore solutions to promote The healthy development of AI technology.